Geometric Inductive Matrix Completion: A hyperbolic approach with unified message passing

Chengkun ZHANG, Hongxu CHEN, Sixiao ZHANG, Guandong XU, Junbin GAO

Research output: Chapter in Book/Report/Conference proceedingChapters

3 Citations (Scopus)

Abstract

Collaborative filtering is a central task in a broad range of recommender systems. As traditional methods train latent variables for user/item individuals under a transductive setting, it requires re-training for out-of-sample inferences. Inductive matrix completion (IMC) solves this problem by learning transformation functions upon engineered features, but it sacrifices model expressiveness and highly depends on feature qualities. In this paper, we propose Geometric Inductive Matrix Completion (GIMC) by introducing hyperbolic geometry and a unified message passing scheme into this generic task. The proposed method is the earliest attempt utilizing capacious hyperbolic space to enhance the capacity of IMC. It is the first work defining continuous explicit feedback prediction within non-Euclidean space by introducing hyperbolic regression for vertex interactions. This is also the first to provide comprehensive evidence that edge semantics can significantly improve recommendations, which is ignored by previous works. The proposed method outperforms the state-of-the-art algorithms with less than 1% parameters compared to its transductive counterparts. Extensive analysis and ablation studies are conducted to reveal the design considerations and practicability for a positive impact to the research community. Copyright © 2022 Association for Computing Machinery.

Original languageEnglish
Title of host publicationProceedings of the 15th ACM International Conference on Web Search and Data Mining, WSDM '22
Place of PublicationNew York
PublisherAssociation for Computing Machinery
Pages1337-1346
ISBN (Electronic)9781450391320
DOIs
Publication statusPublished - Feb 2022

Citation

Zhang, C., Chen, H., Zhang, S., Xu, G., & Gao, J. (2022). Geometric Inductive Matrix Completion: A hyperbolic approach with unified message passing. In Proceedings of the 15th ACM International Conference on Web Search and Data Mining, WSDM '22 (pp. 1337-1346). Association for Computing Machinery. https://doi.org/10.1145/3488560.3498402

Keywords

  • Graph neural networks
  • Inductive matrix completion
  • Hyperbolic space
  • Representation learning

Fingerprint

Dive into the research topics of 'Geometric Inductive Matrix Completion: A hyperbolic approach with unified message passing'. Together they form a unique fingerprint.